Phenotypes for Visual Impairment in the Electronic Health Record

Presentation Time: 03:00 PM - 03:15 PM

Abstract Keywords: Disability, Accessibility, and Human Function, Real-World Evidence Generation, Data Standards
Primary Track: Applications
Programmatic Theme: Clinical Informatics

There are 1.3 billion people worldwide with disabilities and this number is increasing. Patients with disabilities have greater health disparities and lower life expectancies than the general population. Studying these disparities using clinical data in the electronic health record (EHR) has been limited due to the lack disability status designation in the EHR as well as the lack of reliable and validated phenotypes for disabilities. In this study, we compare three different phenotyping methods for visual impairment disability in the EHR: diagnosis codes, visual acuity measurements, and a patient-reported survey. Using data from Oregon Health & Science University, we find that there is very little concordance between these three phenotyping methods and more work is needed to improve phenotypes for visual impairment.

Speaker(s):
Michelle Hribar, PhD
Oregon Health & Science University

Author(s):
Kerry Goetz, MS, PhDc - NIH/NEI; Sally Baxter, MD, MSc - University of California - San Diego; Nicole Weiskopf, PhD - Oregon Health & Science University;